During mammalian embryogenesis and adult tissue homeostasis transforming growth factor .beta. (TGF-.beta.) performs pivotal tasks in intercellular communication (Roberts et al., Growth Factors 8:1-9, 1993). The cellular effects of this pleiotropic factor are exerted by ligand-induced hetero-oligomerization of two distantly related type I and type II serine/threonine kinase receptors, T.beta.R-I and T.beta.R-II, respectively (Lin and Lodish, Trends Cell Biol. 11:972-978, 1993; Derynck, Trends Biochem. Sci. 19-:548-553, 1994; Massague and Weis-Garcia, Cancer Surv. 27:41-64, 1996; ten Dijke et al., Curr. Opin. Cell. Biol. 8:139-145, 1996). The two receptors, which both are required for signalling, act in sequence; T.beta.R-I is a substrate for the constitutively active T.beta.R-II kinase (Wrana et al., Nature 370:341-347, 1994; Weiser et al., EMBO J. 14:2199-2208, 1995).
TGF-.beta. is the prototype of a large family of structurally related proteins that are involved in various biological activities (Massague, et al., Trends Cell Biol. 7:187-192, 1997; Roberts & Sporn, in: Peptide growth factors and their receptors. Part I (Sporn, M. B. and Roberts, A. B., eds) pp. 319-472, Springer-Verlag, Heidelberg (1990); Yingling et al., Biochim. Biophys. Acta 1242:115-136, 1995). The TGF-.beta. "superfamily" includes activins and bone morphogenetic proteins (BMPs) that signal in a similar fashion, each employing distinct complexes of type I and type II serine/threonine kinase receptors (Lin and Lodish, 1993; Derynck, 1994; Massague and Weis-Garcia, 1996; ten Dijke et al., 1996). TGF-.beta. related molecules act in environments where multiple signals interact and are likely to be under tight spatial and chronological regulation. For example, activin and BMP exert antagonistic effects in the development of Xenopus embryos (Graff et al., Cell 85:479-487, 1996). Chordin (Piccolo et al., Cell 86:589-598, 1996) and noggin (Zimmerman et al., Cell 86:599-606, 1996), for example, inhibit the ventralizing effect of BMP4 by binding specifically to the ligand. Likewise, follistatin neutralizes the activity of activin (Hemmati-Brivalou et al., Cell 77:283-295, 1994).
Genetic studies of TGF-.beta.-like signalling pathways in Drosophila and Caenorhabditis elegans have led to the identification of mothers against dpp (Mad) (Sekelsky et al., Genetics 139:1347-1358, 1995) and sma (Savage et al, Proc. Nat. Acad. Sci. USA 93:790-794, 1996) genes, respectively. The products of these related genes perform essential functions downstream of TGF-.beta.-like ligands acting via serine/threonine kinase receptors in these organisms (Wiersdorffet al., Development 122:2153:2163, 1996; Newfeld et al., Development 122:2099-2108, 1996; Hoodless et al., Cell 85:489-500, 1996). Vertebrate homologs of Mad and sma have been termed Smads (Derynck et al., Cell 87:173, 1996) or MADR genes (Wrana and Attisano, Trends Genet. 12:493-496, 1996). Genetic alterations in Smad2 and Smad4/DPC4 have been found in specific tumor subsets, and thus Smads may function as tumor suppressor genes (Hahn et al., Science 271:350-353, 1996; Riggins et al., Nature Genet. 13:347-349, 1996; Eppert et al., Cell 86:543-552, 1996). Smad proteins share two regions of high similarity, termed MH1 and MH2 domains, connected with a variable proline-rich sequence (Massague, Cell 85:947-950, 1996; Derynck and Zhang, Curr. Biol. 6:1226-1229, 1996). The C-terminal part of Smad2, when fused to a heterologous DNA-binding domain, was found to have transcriptional activity (Liu et al., Nature 381:620-623, 1996; Meersseman et al., Mech. Dev. 61:127-1400, 1997). The intact Smad2 protein when fused to a DNA-binding domain, was latent, but transcriptional activity was unmasked after stimulation with ligand (Liu et al., 1996).
Different Smads specify different responses using functional assays in Xenopus. Whereas Smad1 induces ventral mesoderm, a BMP-like response, Smad2 induces dorsal mesoderm, an activin/TGF-.beta.-like response (Graff et al., Cell 85:479-487, 1996; Baker and Harland, Genes & Dev. 10:1880-1889, 1996; Thomsen, Development 122:2359-2366, 1996). Upon ligand stimulation Smads become phosphorylated on serine and threonine residues; BMP stimulates Smad1 phosphorylation, whereas TGF-.beta. induces Smad2 and Smad3 phosphorylation (Hoodless et al., Cell 85:489-500, 1996; Liu et al., 1996; Eppert et al., 1996; Lechleider et al., J. Biol. Chem. 271:17617-17620, 1996; Yingling et al., Proc. Nat'l Aced. Sci. USA93:8940-8944, 1996; Zhang et al., Nature 383:168-172, 1996; Macias-Silva et al., Cell 87:1215-1224, 1996; Nakao et al., J. Biol. Chem. 272:2896-2900, 1996).
Smad4 is a common component of TGF-.beta., activin and BMP signalling (Lagna et al., Nature 383:832-836, 1996; Zhang et al., Curr. Biol. 7:270-276, 1997; de Winter et al., Oncogene 14:1891-1900, 1997). Smad4 phosphorylation has thus far been reported only after activin stimulation of transfected cells (Lagna et al., 1996). After stimulation with TGF-.beta. or activin Smad4 interacts with Smad2 or Smad3, and upon BMP challenge a heteromeric complex of Smad4 and Smad1 has been observed (Lagna et al., 1996). Upon ligand stimulation, Smad complexes translocate from the cytoplasm to the nucleus (Hoodless et al., 1996; Liu et al., 1996; Baker and Harland, 1996; Macias-Silva et al., 1996), where they, in combination with DNA-binding proteins, may regulate gene transcription (Chen et al., Nature 383:691-696, 1996).